Hybrid Data-Expert Explainable Beer Style Classifier

This paper presents a hybrid data-expert approach for building eXplainable Artificial Intelligence (XAI) systems. It combines an opaque machine learning system with several interpretable sys- tems to build a whole XAI system, i.e., a system which provides users with a good interpretability- accuracy trade-off but also with explanation capa- bilities. First, the opaque system acts as an “or- acle” which finds out the most plausible output. Then, the simplest interpretable system carrying out the same output is selected. Finally, a textual explanation of the given output is generated, which emerges as an automatic interpretation of the in- ference process carried out by the selected inter- pretable system. The textual explanation is built by applying a Natural Language Generation Ap- proach. The proposal is validated in a real use case related to beer style classification. 1